Image contour detection is fundamental to many image analysis applications, including image classification. Many conventional image contour detection methods are highly accurate. However, these conventional methods are computationally intensive and require extensive processing time and costs. While these conventional image contour detection methods are sometimes beneficial because they are so precise, not all applications require such precision. Further, the extensive processing time and costs make these computationally intensive image contour detection methods impractical for many image analysis applications, such as image classification of a large collection of images.
Same numbers are used throughout the disclosure and figures to reference like components and features, but such repetition of number is for purposes of simplicity of explanation and understanding, and should not be viewed as a limitation on the various embodiments.